AI Rate Limit Calculator
Turn RPM and TPM limits into your real max throughput.
Providers cap you two ways at once — requests per minute and tokens per minute — and the one that bites depends on how big your calls are. This rpm tpm calculator works out which limit binds first: enter your RPM and TPM caps and the average tokens per request, and it shows your true ceiling in requests per minute plus live headroom gauges. Use it to check openai rate limit headroom or as a requests per minute planner before a batch job, so you scale load without tripping 429s.
Your limits & load
Max throughput
75/min
TPM is the limit that binds first
- RPM allows
- 3,500/min
- TPM allows
- 75/min
- Planned load (TPM)
- 144,000
- Headroom
- 0/min
Your planned load is over the binding limit — batch requests, add backoff, request a higher tier, or shrink each call. Providers apply limits per model and tier and may burst or throttle differently — treat this as a planning gauge.
How it works
- 1
Enter your limits
Put in the requests-per-minute and tokens-per-minute caps for your model and tier, straight from the provider's rate-limit page or the headers on a response.
- 2
Set the request size
Enter the average tokens — input plus output — in a typical call, then how many requests a minute you plan to send. Bigger calls eat the token budget faster.
- 3
Read the binding limit
The headline is your real max throughput. The gauges show how much of each cap your planned load uses, and the tool names whether RPM or TPM runs out first.
Instant & 100% private — nothing is uploaded
Every calculation runs locally in your browser. The prompts, token counts and numbers you enter stay on your own device and are never sent to a server — nothing is stored, logged or shared.
Frequently asked questions
- What's the difference between RPM and TPM?
- RPM caps how many API requests you can make each minute; TPM caps the total tokens — input plus output — you can push through in a minute. You're held to both at once, so your real limit is whichever you hit first for your request size.
- Which limit will I hit first?
- Large requests hit the token limit first: a 90,000 TPM cap at 1,000 tokens a call only allows 90 requests a minute, even if your RPM limit is far higher. Small requests hit the request limit first, since the token budget is nowhere near full. This tool shows which one binds for your numbers.
- How do I get more headroom?
- Shrink each call by trimming the prompt and capping output, batch several jobs into fewer requests, add exponential backoff so retries don't pile on, or request a higher tier from the provider. Raising the limit that binds is what actually helps — lifting the other one does nothing.
- Are these the official limits for my account?
- No. You enter your own caps, since limits vary by provider, model and usage tier and change as your account matures. Check your provider dashboard or a response's rate-limit headers for the current numbers, then plug them in here.
Important
For planning and estimates only. Prices come from a published rate table dated on the page; providers change pricing without notice, and token counts here are approximations. Confirm against the provider’s own pricing before you budget or commit.
More tools
More from the Hivly network
Free sister tools on our other sites.
